Breast cancer incidence is disparate among various populations of women. Moreover, mortality rates do not consistently match incidence. This is particularly striking in African American women, who, at the time of diagnosis often present at a younger age with higher stage, more biologically aggressive tumors. Phenotypically these tumors are highly undifferentiated and aneuploid leading us to believe that these cells have high levels of genetic instability. The centrosome is the accepted intracellular structure responsible for establishment and maintenance of polarity in normal cells, as well as forming mitotic spindle poles, structure critical to stable and accurate chromosome segregation. The hypothesis of this pilot study is that defective centrosomal function is responsible for the aggressive phenotype of certain breast tumors. The study will combine cell and molecular biology to characterize high grade tumors from African American patients, evaluate whether centrosomal anomalies seen in African American women's tumors are present in tumors from other populations, and profile gene expression patterns of these cancers to establish correlative phenotypic and genotypic information. If connections between tumor aggressiveness and centrosomal defects are made, further studies delineating relationships with clinical behavior and epidemiology would be justified. In addition, predictive genetic markers and chemotherapeutic targets may be identified.